Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Advancing the price-performance frontier with GPT‑5.6

openai.com

Post reply on HN
Print view View original

271–280 of 424 posts

Re: Advancing the price-performance frontier with GPT‑5.6

#271

Post by gck1 »

GodelNumbering wrote: post #44

"Half the money I spend on advertising is wasted; the trouble is I don't know which half." -John Wanamaker This applies even more strongly to model choosing. I know for a fact that majority of my work doesn't require a very strong model, but separating the trivial and non-trivial tasks is a famously hard problem (if at all decidable).

Luna is comparable to GPT 5.4 from 4 months ago on many benchmarks. I know many who have said during that time, myself included, that if that's the model they had to use for the rest of their lives, they'd be fine.

GPT 5.4 is/was a very capable model.

Re: Advancing the price-performance frontier with GPT‑5.6

#272

Post by kooi »

vjvjvjvjghv wrote: post #259
jrflo wrote: post #115

Earlier quoted context omitted.

Burning the weights into silicon would be many orders of magnitude increase, not just 10x. It's kind of crazy that this hockey stick the AI hype bros talk about seems more and more every day like it might be real

Not an expert on this but wouldn’t this be possible with something similar to an FPGA?

Weights can be baked into silicone or programmed into hardware ala FPGA, but the context will always be dynamic.

High speed SRAM is where the $$$ is

Re: Advancing the price-performance frontier with GPT‑5.6

#273

Post by bob1029 »

eevmanu wrote: post #232
bob1029 wrote: post #144

Earlier quoted context omitted.

Here is a rough approximation of the pipeline I use: Phase 1 - Run X copies of Luna in parallel over the user's prompt. The purpose is to generate a diverse set of hypotheses. Phase 2 - Run Y copies of Terra in parallel to investigate the hypothesis results, with each receiving them in a randomized order. Phase 3 - Run 1 copy of Sol over investigation reports. The goal is to ensure that the agent covers more initial…

How do you run this phases and parallelization on each? Via just ... "prompting it"? Or do you use any tool in the middle to ensure this agent architecture? Just curious if there is any workflow-like tool in the middle that is helping.

There is a deterministic framework that enforces the process. I use the raw chat completion APIs. No "reasoning effort" is applied on top of these requests (it's not even allowed anymore), so there's actually no way to influence this via prompting.

Re: Advancing the price-performance frontier with GPT‑5.6

#274

Post by bob1029 »

andai wrote: post #220
handfuloflight wrote: post #89

Earlier quoted context omitted.

I definitely think you want the genius model to synthesize everything that rolls up to them.

I think this is an unsolved problem. The most interesting thing I saw here is the Recursive Language Models paper. https://arxiv.org/abs/2512.24601 There's also a great write up here by the author: https://alexzhang13.github.io/blog/2025/rlm/

I used this architecture for a while. The problem I have with it is that starting from one agent and fanning out keeps things mostly aligned with that single reasoning trajectory, even as you get a few layers into the stack. Every recursive invocation is a product of the caller's current state. Diversity doesn't really occur on its own unless the environment (tool calling) is complex/chaotic.

RLM might be more useful on the execution side than on the research side. In fact, these somehow feel like they might be exact inverses of each other in terms of what the ideal architecture looks like. At some point you definitely do need something in the middle that has it all sorted out.

Re: Advancing the price-performance frontier with GPT‑5.6

#275

Post by jaggederest »

vjvjvjvjghv wrote: post #259
jrflo wrote: post #115

Earlier quoted context omitted.

Burning the weights into silicon would be many orders of magnitude increase, not just 10x. It's kind of crazy that this hockey stick the AI hype bros talk about seems more and more every day like it might be real

Not an expert on this but wouldn’t this be possible with something similar to an FPGA?

My understanding is that for FPGA the issue is either it eats all your gates on internal memory if you interleave, or it takes forever to load everything between the SRAM on the board and the actual FPGA component over a bus, last time I looked into it.

Re: Advancing the price-performance frontier with GPT‑5.6

#276

Post by attentive »

pimeys wrote: post #199
GodelNumbering wrote: post #44

"Half the money I spend on advertising is wasted; the trouble is I don't know which half." -John Wanamaker This applies even more strongly to model choosing. I know for a fact that majority of my work doesn't require a very strong model, but separating the trivial and non-trivial tasks is a famously hard problem (if at all decidable).

If you have agents and users, you can run evals and see how far the models go. Luna is not greatest in tool calls, but if you define your problem well and the tools well, it is comparable to Gemini 4 Flash with much lower price tag.

don't expect it to be opus, but luna does coding just fine for its size/price

Re: Advancing the price-performance frontier with GPT‑5.6

#277

Post by Zarathruster »

bakugo wrote: post #51
gentlewater wrote: post #26

Earlier quoted context omitted.

This is gonna put Sonnet 5 in a really awkward spot.

Sonnet and Haiku were already in an awkward spot, likely by design. Anthropic's big marketing push this year has been entirely focused on getting people to use Opus via a Claude Code subscription, to the point that Sonnet is almost viewed as the poor man's alternative, and from what I've seen, almost nobody uses it. Actually, here's an interesting project for all the vibe coders looking for their next front page post…

I've found that Claude nearly always claims to be Opus in the commit message, regardless of the actual model making the commit.

Re: Advancing the price-performance frontier with GPT‑5.6

#278

Post by gck1 »

wmf wrote: post #156
odiroot wrote: post #129

Earlier quoted context omitted.

You just need a very strong frontier model to do triage of your tasks. /s

That's not necessarily a joke; the article proposes exactly that.

It's funny how codex itself can't do Sol orchestrator / Luna implementor out of the box.

Re: Advancing the price-performance frontier with GPT‑5.6

#279

Post by nly »

throwup238 wrote: post #257
8cvor6j844qw_d6 wrote: post #166

Earlier quoted context omitted.

I'd like to imagine the things that can be done with this speed and the current frontier models.

Fully interactive games where you can talk to every NPC by text or voice and have an LLM drive the story (with your own meta prompts to guide it, if you so wish). Maybe even have them generate assets on the fly too. I’m still trying to figure out coding agents. I can’t even begin to imagine the things it would enable. Even the most mundane ideas like LLMs-in-HiFreq-trading have huge implications.

0.03 seconds is an eternity in high frequency trading

You need to be 4 orders of magnitude faster at least

Re: Advancing the price-performance frontier with GPT‑5.6

#280

Post by AaronAPU »

bob1029 wrote: post #273
eevmanu wrote: post #232

Earlier quoted context omitted.

How do you run this phases and parallelization on each? Via just ... "prompting it"? Or do you use any tool in the middle to ensure this agent architecture? Just curious if there is any workflow-like tool in the middle that is helping.

There is a deterministic framework that enforces the process. I use the raw chat completion APIs. No "reasoning effort" is applied on top of these requests (it's not even allowed anymore), so there's actually no way to influence this via prompting.

Does it actually yield results better than just sending it to 5.6 Pro?
Post reply on HN